PodcastSystems Biology

Science Signaling Podcast: 21 December 2010

+ See all authors and affiliations

Science Signaling  21 Dec 2010:
Vol. 3, Issue 153, pp. pc23
DOI: 10.1126/scisignal.3153pc23


This is a conversation with Bernd Bodenmiller and Ruedi Aebersold about a Research Article published in the 21 December 2010 issue of Science Signaling. Bodenmiller and Aebersold describe the effect that targeted removal of individual kinases and phosphatases had on the yeast phosphoproteome. Their study revealed changes in phosphoproteins that were not direct substrates of the targeted enzymes and demonstrated that the network of kinases and phosphatases is surprisingly robust.

(Length: 14 minutes; file size: 8.4 MB; file format: mp3; location: http://podcasts.aaas.org/science_signaling/ScienceSignaling_101221.mp3)

Technical Details

Length: 14 minutes

File size: 8.4 MB

File Format: mp3

RSS Feed: http://stke.sciencemag.org/rss/podcast.xml

Listen to Podcast: http://podcasts.aaas.org/science_signaling/ScienceSignaling_101221.mp3

Educational Details

Learning Resource Type: Audio

Context: High school upper division 11-12, undergraduate lower division 13-14, undergraduate upper division 15-16, graduate, professional, general public & informal education

Intended Users: Teacher, learner

Intended Educational Use: Learn, teach

Discipline: Biochemistry, bioinformatics, cell biology, molecular biology, proteomics

Keywords: Science Signaling, kinase, network analysis, phosphatase, phosphoproteome, proteome, systems biology, yeast


Host – Annalisa VanHookWelcome to the Science Signaling Podcast for December 21st, 2010. I’m Annalisa VanHook, and today I’m speaking with Bernd Bodenmiller and Ruedi Aebersold about a new paper from their group that describes how removing phosphatases one at a time from yeast cells affects the phosphoproteome of the cell (1).

The proteome is the collection of all of the proteins that are present in a cell, tissue, or organism. The phosphoproteome is that subset of thes e proteins that carry a phosphate group. Many cellular processes—including signaling pathways—are regulated by the reversible phosphorylation of proteins. Phosphorylation may control a protein's activity, or determine its binding partners, its localization, or even its lifetime. The phosphoproteome, then, is dynamic—it changes as signaling pathways, metabolism, or other cellular processes change.

In the current study, Bodenmiller, Aebersold, and their colleagues disables most of the yeast kinases and phosphatases one at a time and observed the effect of removing each enzyme on the phosphoproteome. Aebersold explained why they looked at the phosphoproteome rather than just at those proteins that were known to be directly or indirectly regulated by each enzyme.

Interviewee – Ruedi AebersoldIt has been exceedingly difficult to find out a relationship between kinases and phosphatases and their respective substrates. So, if one looks through the literature, there is—in spite of many, many years of research—there is very few confirmed substrates of a particular kinase or phosphatase, and for many kinases or phosphatases, there’s virtually nothing known. And so, through technical advances in the field of proteomics and software engineering, it became now possible to look at a rather large fraction of the phosphoproteome in one single measurement. So, we make no claim that we measure and detect the whole phosphoproteome, but we measure and detect consistently a rather substantial part of it. No one actually knows how big the phosphoproteome is—it is probably quite large, but we do not have the technical capabilities at this point to measure the whole phosphoproteome. But we were assuming that at the level where we can do phosphoproteomic analysis and if we can do it systematically, if we eliminate one particular kinase or phosphatase from the cell, that, among the phosphoproteins or phosphorylation sites that we detect as disappearing when the kinase disappears would be direct substrates. One goal of the study was to piece together, in this stepwise fashion, a network of phosphoproteins and the kinase[s] that phosphorylate them and converse also with the protein phosphatases. So, to basically fill out more of the information in terms of relationships between the enzymes and their substrates, we undertook this study, and this was one of the goals.

Interviewer – Annalisa VanHookBernd, do you have anything to add?

Interviewee – Bernd BodenmillerAlong with what just Ruedi was saying is that so far people always looked at single kinases and phosphatases and tried to define, for example, the direct substrate. But if you think of the kinases and phosphatases as a network, then we always have—or often have—the propagation of a signal. So, if we inhibit or activate one of the kinases or phosphatases, then not only a single protein will be affected, but if the effected protein is another kinase, then this might, in fact, affect further proteins. And therefore, if we want to understand the impact of a given kinase or phosphatase on the cellular signaling network, we actually have to take a global approach to see all the effects on all the different parts of the cell.

Interviewer – Annalisa VanHookDid you identify gross changes in phosphorylation of each protein that you looked at? So, did you look at whether a protein that was unphosphorylated became phosphorylated or a protein that was phosphorylated became unphosphorylated, or did you look at more qualitative changes in that maybe a protein that had two phosphate groups on it before now has four phosphate groups on it? Or a protein that had a phosphate group in one position now had a phosphate group instead in a different amino acid position? Did you look at that level of detail, or were you just looking for an overall change in phosphorylation state?

Interviewee – Bernd BodenmillerSo, what we look ed at is, for a given phosphorylation site, how its abundance quantitatively changed. And so, we therefore can say that a certain phosphopeptide was upregulated, let’s say, three-fold, or in [the] case [that] it’s a direct substrate, it would disappear. And this we can detect with our method.

Interviewer – Annalisa VanHookSo, you weren’t analyzing entire proteins—you were analyzing individual phosphorylation sites?

Interviewee – Bernd BodenmillerPhosphorylated peptides that describe a given phosphorylation site.

Interviewer – Annalisa VanHookRuedi Aebersold

Interviewee – Ruedi AebersoldBut, I think you have a good question, because if a protein has multiple phosphorylation sites, this method allows us to make a statement about the behavior of each phosphorylation site on a particular protein independently. And so, we’re not making a statement just about protein as such, but they’re resolved at the level of each detected phosphorylation site, so they’re treated as independent but, of course, linked back to the overall sequence of the protein.

Interviewer – Annalisa VanHookSome of the phosphatases and the kinases you knocked out by using a drug that inhibits them, and [for] some of them you used mutations. When you analyzed the effect of knocking out these enzymes, were you looking at cells that had grown without the phosphatase or the kinase for some time, and so what you were looking at what essentially their response to the absence of that kinase as they had adapted to function without it? Or were you looking at an immediate response to the removal of the activity of a kinase or a phosphatase?

Interviewee – Bernd BodenmillerSo, in all of the cases where we studied gene deletion mutants…

Interviewer – Annalisa VanHookBernd Bodenmilller.

Interviewee – Bernd Bodenmiller… the cells had ample time to adapt to the missing kinase or phosphatase. And only in those cases where we measured the so-called analog-sensitive kinases—there’s eight of them; so, these are variants of the kinase that can be specifically inhibited—here we measured the changes in the phosphoproteome after [a] short time point.

Interviewer – Annalisa VanHookSo, in the cases where you were removing the function of the enzyme by a genetic mutation, then, the cell had had time to compensate for the lack of that enzyme.

Interviewee – Bernd BodenmillerThat is correct. Yes.

Interviewee – Ruedi AebersoldYes. And actually this was…

Interviewer – Annalisa VanHookRuedi Aebersold.

Interviewee – Ruedi Aebersold…,I think, one of the most surprising outcomes of this study, was exactly how extensively the same, the cell adapts to the absence of one of these enzymes, because in virtually every case where a particular kinase or phosphatase was deleted and the cell had time to adapt, there were a very similar number of phosphorylation events induced as there were reduced. So, these are clearly events where the cell learns to adapt, learns to live without that particular enzyme, and obviously activates specific signaling systems which then lead to the phosphorylation of other proteins or the reduced phosphorylation of a set of proteins which overall leads to a condition for the cell that still allows it to survive. And one of the additional data that are, were generated in the context of this study was to determine with a number of phenotypic parameters how well the cells cope with this adaptation at the level of the phenotype, and this is like growth rate and so on. And one of the interesting observations was that there is many ways the cell can adapt at the molecular level—so, many different phosphorylation patterns and perturbations of these phosphorylation patterns led to rather similar phenotypic appearances.

Interviewer – Annalisa VanHookDid you look at the ability of the cells to respond to different conditions after knocking out a kinase?

Interviewee – Bernd BodenmillerSo, in this study, we only wanted to establish a baseline of what is the reaction of the phosphoproteome or of the system if a certain kinase or phosphatase is removed from the system. And therefore we all put them under the same standard growth conditions, where we had a defined medium.

Interviewee – Ruedi AebersoldBut, of course, the question you ask is a highly interesting one…

Interviewer – Annalisa VanHookRuedi Aebersold.

Interviewee – Ruedi Aebersold…and there are many directions one could take that, and actually some of these studies which are basically follow-up studies on this one presented here are ongoing. But, we certainly do not have the capacity—simply from the measurement and data collection point of view—to take every yeast kinase deletion strain and test it under a whole range of conditions. This would simply be for our capacity too large a study. So, we would then go in a different directions and take maybe a specific set of kinases that one would, for instance, extract from prior knowledge from the literature or from this data set, and then do basically what you suggest—to do knockouts, grow them under various conditions and measure and relate that to the phenotypes.

Interviewer – Annalisa VanHookBernd?

Interviewee – Bernd BodenmillerAnd one thing I would like to add here—because it was also one of the surprising findings—is that we saw for several kinases and phosphatases that, even though it was assumed so far that they have no function under the standard rich growth conditions, we nevertheless saw fingerprint in the phosphoproteome. And this was surprising to see, and this also shows why it was so important first to look at the baseline and then to move on, as Ruedi just said, now to more specific questions.

Interviewer – Annalisa VanHookYour study indicates that the cellular signaling systems in the yeast are very robust. And robustness is a concept that’s talked about in a lot of different contexts in biology. We talk about robustness of developmental pathways, of genetic regulatory pathways, robustness of metabolic pathways, and, in this case, robustness of signaling networks. Robustness generally refers to a system's ability to, to tolerate changes and to compensate. Where does robustness come from, and how do you measure it? Ruedi?

Interviewee – Ruedi AebersoldI think this is an interesting point, and I think this is one of the significant insights from this study is exactly going in that direction. Classically these signaling systems, including those based on reversible phosphorylation, have been viewed as relatively hardwired, linear sequences of events when one enzyme phosphorylates another and then another, and so, a signal progresses usually from the membrane—plasma membrane—eventually into the nucleus. One concept that has been used a lot in that context is the concept of redundancy, because oftentimes it was found by genetic deletion studies that if one knocks out one enzyme, let’s say a kinase—and that’s true for yeast, as well as human cells—that the cell doesn’t really care. It still grows, it still metabolizes, and I think from a biochemical or cell biology point of view what usually is invoked then to explain this situation is the term “redundancy,” which basically indicates that another enzyme is taking over the role of that enzyme that’s been knocked out. And what we arrived at with this study presently is that we don’t think that this idea of redundancy is correct. We get evidence from the data generated in this study from a highly integrated regulatory network consisting of kinases and their substrates. And we think what we’re seeing, if we knock out one gene and not a whole lot is happening at the level of the phenotype, is not due to the fact that simply another enzyme takes over the function of the just-deleted enzyme, but that the system as a whole reacts to compensate. I think this is a fundamental difference of someone basically jumping into the line and filling out the gap to having an effect on the network that’s now measurable, and the network effect as a whole then compensates. And therefore I think we can say that we are dealing with a robust network, because it does compensate to local perturbations, but the way it reacts and compensates is not so simple that just one enzyme takes over the role of the other, but it’s a network effect, which is largely at this point not understood. So, I think the point you raise is a really interesting one, and I think it’s a point of a lot of future research to understand how the system is robust, in a sense, what exactly is happening, and how this robustness is generated.

Interviewer – Annalisa VanHookBernd, Ruedi, thank you for talking with me.

Interviewee – Ruedi AebersoldYou’re welcome. Thanks for the opportunity.

Interviewee – Bernd BodenmillerThank you. It was very nice.

Host – Annalisa VanHookBodenmiller and Aebersold are authors of a Research Article published in the December 21st issue of Science Signaling. That paper is by Bodenmiller and colleagues, and it’s titled, “Phosphoproteomic Analysis Reveals Interconnected System-Wide Responses to Pertrubations of Kinases and Phosphatases in Yeast” (1).


And that concludes this Science Signaling Podcast. If you have any questions or suggestions, please write to us at sciencesignalingeditors@aaas.org. This show is a production of Science Signaling and of AAAS—Advancing Science, Serving Society. I'm Annalisa VanHook, and on behalf of Science Signaling and its publisher, the American Association for the Advancement of Science, thanks for listening.


View Abstract

Related Content

Navigate This Article